| With the rapid development of mobile communication and the continuous improvement of people’s production and living standards,mobile communication systems face important challenges such as increasing user numbers,wider coverage,faster data transmission rates,and more reliable quality of service.In order to solve the above problems,combining the advantages of different architectures and technologies,cell free(CF)large-scale multiple input and multiple output(MIMO)technology emerged as the times require.Unlike traditional cellular architectures,cell free large-scale MIMO systems do not have the concept of cells and cell edges,effectively overcoming the problem of inter cell interference,and being able to provide consistent and efficient data services to users within the region.In this paper,the following research is mainly done under the cell free system architecture.Firstly,the main characteristics of cell free MIMO systems are studied,and the performance comparison between cell free MIMO architectures and traditional cellular MIMO architectures and small cell architectures is presented in detail.The simulation results show that in a single user area equipped with 16 transmitting antennas,the 95%performance bound of the cell free architecture for the reception signal-to-noise ratio(SNR)of the user at any geographical location in the area is improved by 12 d B and 3.5 d B compared to the other two cellular architectures,respectively;In an 8-user area equipped with 16 antennas,the 95%performance bound of the cell free architecture for receiving signal-to-interference noise ratios(SINR)for users at any geographical location in the area has been improved by 20.3 d B and 6.9 d B compared to the other two cellular architectures,respectively.At the same time,this article deduces the data transmission process of the full connected cell free(FC-CF)and the user centered cell free(UC-CF)architecture in detail,and compares the differences between the two.The simulation results show that in a system equipped with 100 APs and 20 users,when using perfect channel estimation,the median uplink and downlink transmission rates of the user centric CF architecture are 15.9 Mbps and 15.8 Mbps higher than those of the fully connected CF architecture,respectively.Secondly,with the goal of maximizing energy efficiency in cell free systems,this paper designs an uplink power control technology based on deep neural networks,and compares it with traditional optimization based algorithms.The simulation results show that in a fully connected cell free network equipped with 100 APs and 40 UEs,the uplink power control technology based on deep neural networks can improve the system energy efficiency by 0.2 Mbits/Joule compared to the uplink power control algorithm based on optimization problem solving,reduce the power consumption of 8Joule,and reduce the computational complexity from(n~3)to o(n~2).At the same time,this paper verifies the reliability of the above conclusions under multiple different system parameter settings.In addition,this paper studies the dynamic user scheduling problem in fully connected cell-free MIMO systems.Concerned about the quality of service of different users with heterogeneous services need in the cell-free MIMO system,this paper proposes a dynamic user scheduling algorithm that considers the use satisfaction rate.By dynamically controlling the number and specific list of user scheduling on different frequency resources in the cell-free system,the system throughput is maximized under the premise of considering the user transmission requirements.Furthermore,a user scheduling transmission scheme based on partial signaling feedback is proposed to balance the contradiction between system load and algorithm performance.Finally,at the end of the article,a summary of the full research and prospects for future research directions are given. |
PDF Full Text Request